OpenDV-2K: Virtual Knot Invariants via 2k-Moves

• OpenDV-2K is a framework for classifying virtual knots by analyzing local 2k-moves and writhe invariants.
• It demonstrates that n-writhe values are preserved modulo k and odd writhe modulo 2k, providing a congruence-based invariant.
• The framework further shows that odd writhe, when combined with Xi-moves, yields a complete classification into exactly k equivalence classes.

In virtual knot theory, a $2k$-move is a local deformation that adds or removes $2k$ half-twists, and the study of its interaction with writhe-type invariants yields a congruence-based framework for comparing virtual knots. Kodai Wada’s paper “Writhes and $2k$-moves for virtual knots” develops this framework by proving that $n$-writhes are preserved modulo $k$ and odd writhe is preserved modulo $2k$ under finite sequences of $2k$-moves, and by characterizing when odd writhe modulo $2k$ completely determines equivalence once $\Xi$-moves are also allowed (Wada, 2023).

1. Local moves and the ambient setting

A $2k$-move is a local move on a knot diagram that adds or removes $2k$0 half-twists, equivalently $2k$1 crossings, in a small region. It generalizes the notion of a crossing change, which is a $2k$2-move. Two virtual knots $2k$3 and $2k$4 are said to be related by a $2k$5-move if there exists a sequence of $2k$6-moves transforming a diagram of $2k$7 into a diagram of $2k$8.

The setting is that of virtual knots, where diagrammatic and Gauss-diagrammatic methods coexist. In the formulation used here, Gauss diagrams provide the combinatorial support for the writhe invariants. The paper’s central question is not merely whether a given invariant is unchanged by a $2k$9-move, but whether its value is preserved only up to congruence, and whether such congruence data is sufficient for classification in the presence of additional local moves.

A second local operation, the $2k$0-move, is a rearrangement that exchanges the positions of the first and third crossings among three consecutive real crossings. In the paper, this move is tied specifically to the behavior of odd writhe and to the classification theorem modulo $2k$1.

2. Writhe invariants from Gauss diagrams

For a virtual knot $2k$2, the $2k$3-writhe $2k$4 is an integer-valued invariant defined via its Gauss diagram. Each chord in the diagram carries a sign and an index, where the index is the sum of signed endpoints between the endpoints of the chord. Then $2k$5 is the sum of the signs of all chords with index $2k$6. For $2k$7, this is a knot invariant.

The family $2k$8 generalizes writhe by decomposing crossing data according to chord index rather than aggregating all signed crossings into a single count. This index-sensitive structure is what makes congruence statements under $2k$9-moves meaningful: the move does not preserve the integers $n$0 literally in general, but it preserves their residue classes modulo $n$1.

The odd writhe $n$2 is defined by summing the $n$3-writhes over odd indices:

$n$4

This invariant was introduced by Kauffman and functions as a particularly effective tool for distinguishing virtual knots. In Wada’s treatment, odd writhe is the invariant that survives not just as an obstruction, but as a complete classifier for the equivalence relation generated by $n$5-moves together with $n$6-moves.

3. Congruence under $n$7-moves

The basic congruence theorem states that if two virtual knots $n$8 and $n$9 are related by a finite sequence of $k$0-moves, then the following hold (Wada, 2023):

1. For every nonzero integer $k$1,

$k$2

2. For odd writhe,

$k$3

These statements isolate the precise arithmetic stability induced by the move. The invariant values need not agree as integers, but their residue classes are fixed under the generated equivalence relation. In particular, the collection of residues of all $k$4-writhes modulo $k$5, together with the residue of odd writhe modulo $k$6, gives a strong obstruction: if two virtual knots have different values, they are in different $k$7-move classes.

The following summary records the preservation properties stated in the paper.

Invariant	Preserved under $k$8-moves	Congruence modulus
$k$9 for $2k$0	Yes	$2k$1
$2k$2	Yes	$2k$3

A common simplification is to speak as though odd writhe alone classifies $2k$4-move equivalence. The theorem above does not say that. It gives a necessary condition for $2k$5-move equivalence, and it shows that odd writhe is preserved modulo $2k$6 under that relation. Completeness requires the introduction of $2k$7-moves.

4. The $2k$8-move and the classification theorem

The paper’s main classification theorem gives a necessary and sufficient condition for agreement of odd writhe modulo $2k$9 (Wada, 2023). For two virtual knots $2k$0 and $2k$1, the following are equivalent:

1.

$2k$2

1. $2k$3 and $2k$4 are related by a finite sequence of $2k$5-moves and $2k$6-moves.

This equivalence is sharper than the obstruction theorem. It identifies odd writhe modulo $2k$7 as a complete invariant not for $2k$8-moves alone, but for the larger equivalence relation generated jointly by $2k$9-moves and $2k$0-moves. Thus the classification problem changes once $2k$1-moves are admitted: the full family of $2k$2-writhes modulo $2k$3 is no longer needed, and the odd writhe residue class becomes sufficient.

The distinction is conceptually important. The invariants $2k$4 encode more refined index data, while the odd writhe packages only the odd-index contributions. The theorem shows that, after enlarging the move set by $2k$5-moves, this coarser datum exactly matches the resulting equivalence relation. A plausible implication is that $2k$6-moves erase some of the finer index-level information that remains visible under pure $2k$7-move equivalence.

5. Representative systems and normal forms

A corollary of the classification theorem gives a complete representative system for equivalence classes under $2k$8-moves and $2k$9-moves (Wada, 2023):

$\Xi$0

where $\Xi$1 is a virtual knot with odd writhe

$\Xi$2

Accordingly, there are exactly $\Xi$3 equivalence classes of virtual knots up to $\Xi$4-moves and $\Xi$5-moves, indexed by

$\Xi$6

This transforms the classification problem into a finite arithmetic one: each class is determined by the residue of odd writhe modulo $\Xi$7, represented by an even integer $\Xi$8 with $\Xi$9.

The paper also describes a concrete Gauss-diagram model. The Gauss diagram $2k$0 represents a virtual knot $2k$1 with $2k$2 shell-pairs, and its odd writhe is $2k$3. Moreover, all virtual knots can be moved, by $2k$4-moves and $2k$5-moves, into one of the knots $2k$6 with $2k$7. This provides an explicit normal-form picture for the equivalence classes, rather than only an abstract classification by congruence.

6. Consequences, examples, and scope

Several implications are stated explicitly in the paper’s summary material (Wada, 2023). First, the congruence classes of $2k$8-writhes modulo $2k$9 and odd writhe modulo $2k$00 furnish an obstruction to $2k$01-move equivalence: distinct values force distinct classes. Second, for the enlarged equivalence generated by $2k$02-moves and $2k$03-moves, the classification collapses to odd writhe modulo $2k$04 alone.

The paper also records an unknotting-type statement: flat $2k$05-moves are sufficient to trivialize any flat virtual knot. In the source summary this is said to echo a role analogous to crossing changes in classical knot theory, while remaining specific to the virtual setting. This places $2k$06-moves in a broader structural context: they are not auxiliary bookkeeping operations, but moves with strong simplifying power.

A further statement concerns distance realization: for any virtual knot $2k$07 and positive integer $2k$08, there is a virtual knot $2k$09 with $2k$10-move distance exactly $2k$11 from $2k$12. This indicates that the move metric is nontrivial at all positive scales. The source does not elaborate the construction in detail, but the existence statement situates $2k$13-moves not only as equivalence generators but also as a basis for a meaningful quantitative notion of separation.

One potential misconception is that the results reduce virtual-knot classification broadly to odd writhe. The precise statement is narrower. Odd writhe modulo $2k$14 is complete only for the equivalence relation generated by both $2k$15-moves and $2k$16-moves. For $2k$17-moves by themselves, the paper establishes congruence invariance and obstruction results, not a complete classification.

7. Position within virtual-knot invariant theory

The paper’s contribution lies in linking local move theory with index-sensitive Gauss-diagram invariants. The $2k$18-writhes $2k$19, defined chordwise by sign and index, behave arithmetically under $2k$20-moves: each is preserved modulo $2k$21, while their odd-index aggregate $2k$22 is preserved modulo $2k$23. The main theorem then shows that the odd writhe residue class is exactly the datum needed once $2k$24-moves are included (Wada, 2023).

This yields two complementary viewpoints on virtual-knot comparison. From one viewpoint, the family of all $2k$25-writhes detects obstructions to pure $2k$26-move equivalence. From the other, the odd writhe alone organizes the combined $2k$27-move/$2k$28-move classification into exactly $2k$29 classes represented by $2k$30. The result is a compact arithmetic description of a move-generated equivalence relation on virtual knots, grounded in explicit Gauss-diagram invariants and representative models.